You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: docs/README.md
+25-21Lines changed: 25 additions & 21 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -1,14 +1,18 @@
1
-
# Welcome to Context Engineering Kit
1
+
---
2
+
icon: brain-circuit
3
+
---
4
+
5
+
# Context Engineering Kit
2
6
3
7
The Context Engineering Kit (CEK) is a curated marketplace of advanced context engineering techniques and patterns designed specifically for Claude Code. It provides prompts for extensibly tested and benchmarked techniques that enhance LLM output quality, speficially focusing on code genration, research and problem solving.
4
8
5
9
## Key Features
6
10
7
-
-**Simple to Use** - Easy to install and use without any dependencies. Contains automatically used skills and self-explanatory commands.
8
-
-**Token-Efficient** - Carefully crafted prompts and architecture, preferring commands over skills, to minimize populating context with unnecessary information.
9
-
-**Quality-Focused** - Each plugin is focused on meaningfully improving agent results in a specific area.
10
-
-**Granular** - Install only the plugins you need. Each plugin loads only its specific agents, commands, and skills. Each without overlap and redundant skills.
11
-
-**Scientifically proven** - Plugins are based on proven techniques and patterns that were tested by well-trusted benchmarks and studies, with exception to development workflows that based on popular projects and frameworks.
11
+
***Simple to Use** - Easy to install and use without any dependencies. Contains automatically used skills and self-explanatory commands.
12
+
***Token-Efficient** - Carefully crafted prompts and architecture, preferring commands over skills, to minimize populating context with unnecessary information.
13
+
***Quality-Focused** - Each plugin is focused on meaningfully improving agent results in a specific area.
14
+
***Granular** - Install only the plugins you need. Each plugin loads only its specific agents, commands, and skills. Each without overlap and redundant skills.
15
+
***Scientifically proven** - Plugins are based on proven techniques and patterns that were tested by well-trusted benchmarks and studies, with exception to development workflows that based on popular projects and frameworks.
12
16
13
17
## IDEs and CLIs support
14
18
@@ -18,37 +22,37 @@ Currenty this project support only Claude Code CLI, but we plan to support other
18
22
19
23
Start here to get up and running quickly:
20
24
21
-
-[Getting Started](getting-started.md) - Installation, setup, and your first plugin
22
-
-[User Guide](user-guide.md) - Common workflows and usage patterns
Copy file name to clipboardExpand all lines: docs/getting-started.md
+34-25Lines changed: 34 additions & 25 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -1,14 +1,19 @@
1
-
# Getting Started with Context Engineering Kit
1
+
---
2
+
description: >-
3
+
This guide will help you install Context Engineering Kit to your Claude Code
4
+
and start using plugins.
5
+
icon: rocket
6
+
---
2
7
3
-
This guide will help you install Context Engineering Kit to your Claude Code and start using plugins.
8
+
# Getting Started
4
9
5
10
## Prerequisites
6
11
7
12
Before you begin, ensure you have:
8
13
9
14
**Claude Code installed** - The official CLI tool from Anthropic
10
-
- If not installed, visit [Claude Code documentation](https://docs.anthropic.com/claude/docs/claude-code) for installation instructions
11
15
16
+
* If not installed, visit [Claude Code documentation](https://docs.anthropic.com/claude/docs/claude-code) for installation instructions
12
17
13
18
## Quick Start
14
19
@@ -27,10 +32,11 @@ Then add the Context Engineering Kit marketplace to make all plugins available:
27
32
```
28
33
29
34
**What happens:**
30
-
- The marketplace metadata is downloaded and cached locally
31
-
- All available plugins become visible in your plugin list
32
-
- No plugins are installed yet - this only makes them available
33
-
- No agents, commands, or skills are loaded - your context remains clean
35
+
36
+
* The marketplace metadata is downloaded and cached locally
37
+
* All available plugins become visible in your plugin list
38
+
* No plugins are installed yet - this only makes them available
39
+
* No agents, commands, or skills are loaded - your context remains clean
34
40
35
41
**Verify it worked:**
36
42
@@ -49,9 +55,10 @@ We recommend starting with the **Reflexion plugin** - it introduces feedback and
49
55
```
50
56
51
57
**What happens:**
52
-
- The Reflexion plugin is installed in your Claude Code environment
53
-
- Three new commands become available: `/reflexion:reflect`, `/reflexion:memorize`, `/reflexion:critique`
54
-
- Plugin-specific skills and agents are loaded into Claude's context in future sessions
58
+
59
+
* The Reflexion plugin is installed in your Claude Code environment
60
+
* Three new commands become available: `/reflexion:reflect`, `/reflexion:memorize`, `/reflexion:critique`
61
+
* Plugin-specific skills and agents are loaded into Claude's context in future sessions
55
62
56
63
### Step 3: Use Your First Command
57
64
@@ -67,23 +74,26 @@ Claude will provide an initial response. Now, use Reflexion to ask it to reitera
67
74
/reflexion:reflect
68
75
```
69
76
70
-
**What happens:**
71
-
Claude reviews its previous response using self-refinement techniques, identifies areas for improvement, and generates an enhanced version with deeper analysis.
77
+
**What happens:** Claude reviews its previous response using self-refinement techniques, identifies areas for improvement, and generates an enhanced version with deeper analysis.
72
78
73
-
**Expected result:**
74
-
Claude will analyze its previous response critically, identify specific improvements (e.g., "I should have considered error propagation patterns"), and provide an enhanced response with more detail and better recommendations.
79
+
**Expected result:** Claude will analyze its previous response critically, identify specific improvements (e.g., "I should have considered error propagation patterns"), and provide an enhanced response with more detail and better recommendations.
75
80
76
81
**Try the memorize command:**
77
82
78
83
```bash
79
84
/reflexion:memorize
80
85
```
81
86
82
-
**What happens:**
83
-
Claude identifies key learnings from the interaction, updates your project's `CLAUDE.md` file with curated insights, and builds a knowledge base that future Claude sessions can leverage.
87
+
**What happens:** Claude identifies key learnings from the interaction, updates your project's `CLAUDE.md` file with curated insights, and builds a knowledge base that future Claude sessions can leverage.
84
88
85
89
## What's Next?
86
90
91
+
*[**User Guides**](guides/) - Complete guides to using the marketplace
Practical how-to guides for getting the most out of the Context Engineering
4
+
Kit plugins.
5
+
icon: book
6
+
---
2
7
3
-
Practical how-to guides for getting the most out of the Context Engineering Kit plugins.
8
+
# Guides
4
9
5
10
## Workflow Guides
6
11
7
-
| Guide | Description |
8
-
|-------|-------------|
9
-
|[Feature Development](./feature-development.md)| Complete feature development with reflection and code review quality gates |
10
-
|[Spec-Driven Development](./spec-driven-development.md)| Full SDD workflow for complex features requiring planning |
11
-
|[Bug Investigation](./bug-investigation.md)| Systematic bug fixing with root cause analysis |
12
-
|[Code Quality Improvement](./code-quality-improvement.md)| Systematic quality improvements using Kaizen methodology |
13
-
|[Custom Extensions](./custom-extensions.md)| Build project-specific commands and skills |
14
-
|[Project Setup](./project-setup.md)| Initialize new projects with best practices |
15
-
|[PR Review](./pr-review.md)| Comprehensive pull request review process |
16
-
|[Brainstorming to Implementation](./brainstorming-to-implementation.md)| Transform vague ideas into working features |
12
+
<table><thead><tr><th width="279">Guide</th><th>Description</th></tr></thead><tbody><tr><td><a href="feature-development.md">Feature Development</a></td><td>Complete feature development with reflection and code review quality gates</td></tr><tr><td><a href="spec-driven-development.md">Spec-Driven Development</a></td><td>Full SDD workflow for complex features requiring planning</td></tr><tr><td><a href="bug-investigation.md">Bug Investigation</a></td><td>Systematic bug fixing with root cause analysis</td></tr><tr><td><a href="code-quality-improvement.md">Code Quality Improvement</a></td><td>Systematic quality improvements using Kaizen methodology</td></tr><tr><td><a href="custom-extensions.md">Custom Extensions</a></td><td>Build project-specific commands and skills</td></tr><tr><td><a href="project-setup.md">Project Setup</a></td><td>Initialize new projects with best practices</td></tr><tr><td><a href="pr-review.md">PR Review</a></td><td>Comprehensive pull request review process</td></tr><tr><td><a href="brainstorming-to-implementation.md">Brainstorming to Implementation</a></td><td>Transform vague ideas into working features</td></tr></tbody></table>
17
13
18
14
## Quick Reference
19
15
20
16
### Daily Development
21
-
-**New feature**: [Feature Development](./feature-development.md) or [Spec-Driven Development](./spec-driven-development.md) for complex features
0 commit comments